Magnetic resonance imaging in children with implants.

As access to MRI in pediatrics increases, the radiologist needs to become acquainted with the basic principles of MRI safety. As part of the image acquisition, the static magnetic field, gradient system, and the radiofrequency transmit-receive coil interact with medical and non-medical implants and can result in serious injury. The main stage of risk triage is based on the determination of whether the implant is MRI-safe, conditional, unsafe or unknown. Guiding principles include the strict adherence to manufacturer specifications for MRI-conditional implants and the assumption that an unknown implant is MR-unsafe. In this article we review considerations for common medical implants encountered in pediatrics including ventriculoperitoneal shunts, orthopedic hardware, orthodontic hardware, pacemakers, vascular stents, vagal nerve stimulators and cochlear implants. Finally, we review a set of high-yield considerations, including the non-communicative patient (sedated or non-verbal), susceptibility artifacts from unclear source, and the approach to an unknown implant.

Neck CT angiography examinations for pediatric oropharyngeal trauma: diagnostic yield and proposal of a new targeted technique.

BACKGROUND: Neck computed tomography (CT) angiography is commonly ordered for pediatric patients with soft palate trauma to exclude vascular injury. Debate exists regarding what type of imaging is indicated in this setting, particularly amid growing concern that standard neck CT angiography results in considerable radiation exposure. OBJECTIVE: To assess the diagnostic yield and estimated dose reduction of a novel targeted protocol extending from the skull base to the hyoid bone to evaluate pediatric oropharyngeal trauma. MATERIALS AND METHODS: A retrospective imaging and medical chart review was performed of patients for whom a neck CT angiography was obtained for an indication of oropharyngeal trauma between 2008 and 2018. Effective dose and size-specific dose estimates (SSDEs) were estimated for standard and targeted neck CT angiography protocols with calculation of percent dose reduction of the targeted exams. RESULTS: Ninety-eight CT angiography examinations were reviewed. No cases were positive for neurological or major vessel injury; one case was positive for small vessel extravasation. Clinically significant nonvascular findings included phlegmonous change, retained foreign body, retropharyngeal/mediastinal air and pterygoid process fracture. With the exception of mediastinal air, all findings would have been included in the targeted protocol. Effective dose and SSDE were calculated for all cases where CTDIvol (volume CT dose index) had been reported (n=72). There was a statistically significant reduction in dose for the targeted protocol with an effective dose decrease of 69.7%±10.5% (P=0.009) and SSDE decrease of 53.9%±14.7% (P=0.01). Limiting ionizing radiation to the lung apices, esophagus and thyroid gland provided the greatest dose savings. CONCLUSION: Based on low diagnostic yield and high radiation dose associated with standard neck CT angiography for evaluating oropharyngeal trauma, a targeted protocol is recommended, resulting in significantly less dose to the neck, while preserving diagnostic yield.

Placental Pathology in Neonatal Stroke: A Retrospective Case-Control Study.

OBJECTIVE: To assess the association of placental abnormalities with neonatal stroke. STUDY DESIGN: This retrospective case-control study at 3 academic medical centers examined placental specimens for 46 children with neonatal arterial or venous ischemic stroke and 99 control children without stroke, using a standard protocol. Between-group comparisons used χ2 and Fisher exact t test. Correlations used Spearman correlation coefficient. RESULTS: Case placentas were more likely than controls to meet criteria for ≥1 of 5 major categories of pathologic abnormality (89% vs 62%; OR, 5.1; 95% CI, 1.9-14.0; P = .0007) and for ≥2 categories (38% vs 8%; OR, 7.3; 95% CI, 2.9-19.0; P < .0001). Fetal vascular malperfusion occurred in 50% of cases and 17% of controls (OR, 4.8; 95% CI, 2.2-10.5; P = .0001). Amniotic fluid inflammation occurred in 46% of cases with arterial ischemic stroke vs 25% of controls (OR, 2.6; 95% CI, 1.1-6.1; P = .037). There was evidence of a "stress response" (meconium plus elevated nucleated red blood cells) in 24% of cases compared with 1% of controls (OR, 31; 95% CI, 3.8-247.0; P < .0001). CONCLUSIONS: Placental abnormality was more common in children with neonatal stroke compared with controls. All placental findings represent subacute-to-chronic intrauterine stressors. Placental thrombotic processes were associated with both arterial and venous stroke. Our findings provide evidence for specific mechanisms that may predispose to acute perinatal stroke. Amniotic fluid inflammation associated with neonatal arterial ischemic stroke deserves further investigation.

Workup for Perinatal Stroke Does Not Predict Recurrence.

BACKGROUND AND PURPOSE: Perinatal stroke, including neonatal and presumed perinatal presentation, represents the age in childhood in which stroke occurs most frequently. The roles of thrombophilia, arteriopathy, and cardiac anomalies in perinatal ischemic stroke are currently unclear. We took a uniform approach to perinatal ischemic stroke evaluation to study these risk factors and their association with recurrent stroke. METHODS: We reviewed records of perinatal stroke patients evaluated from August 2008 to February 2016 at a single referral center. Demographics, echocardiography, arterial imaging, and thrombophilia testing were collected. Statistical analysis was performed using Fisher exact test. RESULTS: Across 215 cases, the median follow-up was 3.17 years (1.49, 6.46). Females comprised 42.8% of cases. Age of presentation was neonatal (110, 51.2%) or presumed perinatal (105, 48.8%). The median age at diagnosis was 2.9 days (interquartile range, 2.0-9.9) for neonatal stroke and 12.9 months (interquartile range, 8.7-32.8) for presumed perinatal stroke. Strokes were classified as arterial (149, 69.3%), venous (60, 27.9%), both (4, 1.9%), or uncertain (2, 0.9%) by consensus imaging review. Of the 215 cases, there were 6 (2.8%) recurrent ischemic cerebrovascular events. Abnormal thrombophilia testing was not associated with recurrent stroke, except for a single patient with combined antithrombin deficiency and protein C deficiency. After excluding venous events, 155 patients were evaluated for arteriopathy and cardioembolic risk factors; neither was associated with recurrent stroke. Positive family history of thrombosis was not predictive of abnormal thrombophilia testing. CONCLUSIONS: Thrombophilia, arteriopathy, or cardioembolic risk factors were not predictive of recurrent events after perinatal stroke. Thrombophilia evaluation in perinatal stroke should only rarely be considered.

Predictors of Stroke After Transient Ischemic Attack in Children.

BACKGROUND AND PURPOSE: Transient ischemic attack (TIA) in children has received far less attention compared with TIA in adults. The risk factors of stroke after TIA in children are relatively unknown. We aimed to determine the percentage of children who have stroke after TIA and the risk factors associated with stroke after TIA. METHODS: We searched the medical records at Boston Children's Hospital for the year 2010 to find children who were evaluated for TIA to determine associated risk factors of stroke after TIA. We included children who were evaluated in 2009 through 2010 for TIA and had magnetic resonance imaging. We examined follow-up imaging through August 2014 for subsequent stroke. Logistic regression was used to calculate odds ratios for factors in our cohort who are associated with stroke after presentation with TIA. RESULTS: We identified 63 children who experienced a TIA. The mean time of imaging follow-up was 4.5 years after TIA presentation. Of the 63 children, 10 (16%) developed radiological evidence of ischemic cerebral injury within the follow-up period. Four of the 10 (6%) demonstrated diffusion abnormalities on magnetic resonance imaging at TIA presentation, whereas 8 (13%) had a stroke after their TIA. Arteriopathy, female sex, and autoimmune disorders were significantly associated with stroke after TIA. CONCLUSIONS: In our cohort of children, stroke occurred after TIA at a rate similar to that seen in adults, but the risk factors for stroke after TIA in children are different.

Retroclival collections associated with abusive head trauma in children.

Retroclival collections are rare lesions reported almost exclusively in children and strongly associated with trauma. We examine the incidence and imaging characteristics of retroclival collections in young children with abusive head trauma. We conducted a database search to identify children with abusive head trauma ≤ 3 years of age with brain imaging performed between 2007 and 2013. Clinical data and brain images of 65 children were analyzed. Retroclival collections were identified in 21 of 65 (32%) children. Ten (48%) were subdural, 3 (14%) epidural, 2 (10%) both, and 6 (28%) indeterminate. Only 8 of 21 retroclival collections were identifiable on CT and most were low or intermediate in attenuation. Eighteen of 21 retroclival collections were identifiable on MRI: 3 followed cerebral spinal fluid in signal intensity and 15 were bloody/proteinaceous. Additionally, 2 retroclival collections demonstrated a fluid-fluid level and 2 enhanced in the 5 children who received contrast material. Sagittal T1-weighted images, sagittal fluid-sensitive sequences, and axial FLAIR (fluid-attenuated inversion recovery) images showed the retroclival collections best. Retroclival collections were significantly correlated with supratentorial and posterior fossa subdural hematomas and were not statistically correlated with skull fracture or parenchymal brain injury. Retroclival collections, previously considered rare lesions strongly associated with accidental injury, were commonly identified in this cohort of children with abusive head trauma, suggesting that retroclival collections are an important component of the imaging spectrum in abusive head trauma. Retroclival collections were better demonstrated on MRI than CT, were commonly identified in conjunction with intracranial subdural hematomas, and were not significantly correlated with the severity of brain injury or with skull fractures.

Adverse events in pediatric cochlear implant patients undergoing magnetic resonance imaging.

OBJECTIVE: To investigate the prevalence and nature of adverse events in magnetic resonance imaging (MRI) of pediatric cochlear implant (CI) patients. METHODS: Retrospective chart review at a tertiary pediatric hospital. CI patients who underwent MRI from 2004 through 2019 were identified via our internal radiology database. Comorbidities, CI model, age at MRI, number of MRIs, type of MRIs, indication for MRIs, precautions taken for MRIs, quality of MRIs, anesthesia during MRIs, patient language abilities, and adverse events were recorded from the electronic medical record. The literature was reviewed, and our results were compared to those of previous similar series. RESULTS: From 2004 to 2019, 12 pediatric patients (17 ears) with CIs underwent 22 MRIs. 12 MRIs were performed in CI patients with retained internal magnet. 4/22 MRIs resulted in morbidity; 2 patients experienced pain requiring MRI abortion, 1 experienced magnet rotation requiring surgical replacement, and 1 underwent operative removal of the magnet prior to the scan with surgical replacement thereafter. 19/22 MRIs were performed to evaluate the brain; 17/22 of the radiologic reports noted limitation of evaluation due to artifact. 18/22 MRIs required the administration of anesthesia. 9 of the 22 MRI events involved 2 patients whose CIs had been without internal magnet in anticipation of future MRI requirement. CONCLUSIONS: Adverse events affecting pediatric patients with CI can occur as a result of MRI, despite appropriate precautions. Safety requires consideration of factors unique to a pediatric hearing-impaired population. Clinicians must remain informed on best practices and manufacturer recommendations.

Pathways for Neuroimaging of Neonatal Stroke.

PURPOSE: To provide consensus-based, suggested imaging protocols to facilitate the accurate and timely diagnosis of a neonate with symptoms concerning for stroke. METHODS: The Writing Group, an international collaboration of pediatric neurologists and neuroradiologists with expertise in perinatal and childhood stroke, participated in a series of pediatric stroke neuroimaging symposia. These discussions, in conjunction with extensive literature review, led to a consensus for imaging protocols to guide practitioners in the diagnosis of neonatal stroke subtypes as defined by the National Institute of Neurological Disorders and Stroke Common Data Elements. The epidemiology, clinical presentation, and associated risk factors for arterial ischemic stroke, cerebral sinovenous thrombosis, and hemorrhagic stroke are reviewed, with a focused discussion regarding the role of neuroimaging for each subtype. RESULTS: In a neonate with suspected stroke, magnetic resonance imaging is the preferred modality, given the lack of X-irradiation, superior anatomic resolution, and sensitivity for acute ischemia. Core recommended sequences include diffusion-weighted imaging and apparent diffusion coefficient mapping to diagnose acute ischemia, gradient-recalled echo or susceptibility-weighted imaging to detect intracranial blood and its breakdown products, and T1- and T2-weighted imaging to assess for myelination, extra-axial blood, and edema. Magnetic resonance angiography of the brain may be useful to detect vascular abnormalities, with venography if venous sinus thrombosis is suspected. The application of more novel sequences, as well as the utility of follow up-imaging, is also discussed.

Calvarial defects and skeletal dysplasia in patients with neurofibromatosis Type 1.

OBJECT: Skull defects, including sphenoid dysplasia and calvarial defects, are rare but distinct findings in patients with neurofibromatosis Type 1 (NF1). The underlying pathophysiology is unclear. The goal of this study was to identify the clinical characteristics and natural history of skull defects in patients with NF1. METHODS: An electronic search engine of medical records was used to identify patients with NF1 and bony skull anomalies. All clinical, radiographic, pathology, and operative reports were reviewed. The relationship between bony anomalies and significant clinical associations was evaluated. This study received institutional review board approval. RESULTS: Twenty-one patients were identified. The mean age at NF1 diagnosis was 4.2 years. The mean age at skull defect diagnosis was 8.8 years (9.7 years in the sphenoid wing dysplasia group and 11.9 years in the calvarial defect group). Sphenoid dysplasia was associated with a plexiform neurofibroma or dural ectasia in 73.3% and 80.0% of cases, respectively. Calvarial defects were associated with a plexiform neurofibroma or dural ectasia in 66.7% and 33.3% of patients, respectively. An absence of either an associated neurofibroma or ectasia was not noted in any patient with sphenoid wing dysplasia or 25.0% of those with calvarial defects. In 6 patients, both types of skull defects presented simultaneously. Serial imaging studies were obtained for a mean follow-up time of 7.5 years (range 0.4-20.0 years). Of these patients with serial imaging, radiographic progression was found in 60% of cases of calvarial defects and 56% of cases of sphenoid wing dysplasia. Two patients underwent surgical repair of a skull defect, and both required repeat procedures. CONCLUSIONS: The majority of skull defects in patients with NF1 were associated with an adjacent structural lesion, such as a plexiform neurofibroma or dural ectasia. This findings from this cohort also support the concept of progression in defect size in more than half of the patients. Potential mechanisms by which these secondary lesions contribute to pathogenesis of the bony defect may include changes in the bony microenvironment. A better understanding of the pathophysiology of skull defects will help guide detection, improve treatment and outcome, and may contribute to the understanding of the pathogenesis of bony lesions in NF1.